Skip to content
OSRobotics

HumanML3D

Human motions paired with text descriptions for language-to-motion.

HumanML3D annotates 14,616 AMASS motions with 44,970 text descriptions. It underlies text-to-motion models whose outputs are increasingly retargeted onto humanoid robots.

Description

HumanML3D annotates 14,616 AMASS motions with 44,970 text descriptions. It underlies text-to-motion models whose outputs are increasingly retargeted onto humanoid robots.